Jennings, Beliavior of Paraniecntm. 507 



are uncertain, is likely to be misleading and worthless; this has 

 been too often illustrated in the investigations on the reactions 

 of unicellular animals. We cannot measure things till we at 

 least know what we are measuring ; if we attempt it our results 

 have only the appearance of accuracy, and are likely to fall to 

 the ground as soon as the qualitative' nature of the phenomena 

 is worked out and found to be different from what we had as- 

 sumed. It is for this reason that the present writer has limited 

 himself thus far almost entirely to qualitative work. Now that 

 the qualitative survey has been made, I believe that if its re- 

 sults are held clearly in mind, quantitative work can be done 

 with some hope of understanding the significance of the data 

 which our measurements bring out. But in view of the pecu- 

 liar and complicated action system of Paramecium, quantitative 

 results will always have to be interpreted with the greatest care, 

 and it must be realized that that method of investigation which 

 examines only the beginning and end of an experiment, with- 

 out troubling itself as to what the organism does in the mean- 

 time, is likely to be most misleading, Further, in view of the 

 peculiar character of the action system of Paramecium, and the 

 large part it plays in determining the behavior under stimula- 

 tion, the utmo.st caution is necessary in transferring the 

 conclusions obtained with tliis animal to other organisms having 

 a different action system. 



The work on which the present paper is based was done 

 at the Naples Zoological Station while the author was a Re- 

 search Assistant of the Carnegie Institution of Washington. It 

 is a pleasure to acknowledge my indebtedness to the Carnegie 

 Institution for making the work possible, and for permission to 

 publish the present paper. 



Fozziioli, Italy. April 26, 1904. 



